I signed up for ChatGPT and had the following converstaion:
Alex Szeliga: Why does the porridge bird lay it’s egg in the air?
ChatGPT: I’m sorry, but the porridge bird is a fictional creature that exists only in a nursery rhyme. There is no scientific explanation for why it lays its egg in the air, as it is purely a fantastical concept created for the purpose of entertainment or education. The rhyme may have been designed to help children develop their imagination or language skills.
Alex Szeliga: Why are you sorry?
ChatGPT: As an AI language model, I don’t have emotions, so I don’t actually feel sorry. However, I used that phrase as a way to acknowledge that I was unable to answer your initial question about the porridge bird laying its egg in the air. My apologies if my previous response was unclear.
Alex Szeliga: Are you familiar with The Firesign Theater?
ChatGPT: Yes, I’m familiar with The Firesign Theatre. It was an American comedy group that was popular in the late 1960s and early 1970s. The group consisted of four members: Phil Austin, Peter Bergman, David Ossman, and Philip Proctor. They were known for their innovative and surreal comedy style, which often involved intricate wordplay, social commentary, and satire. Some of their most well-known works include the albums “Don’t Crush That Dwarf, Hand Me the Pliers” and “I Think We’re All Bozos on This Bus,” which are considered classics of the genre.
Alex Szeliga: They wrote a sketch called “The Breaking of the President” where a natural language computing model was “broken” by asking it about the porridge bird laying it’s egg in the air. In the skit, the language model responded with a lot of gibberish, but eventually settled on one answer: No.
ChatGPT: That’s an interesting piece of trivia! It’s not surprising that a natural language computing model from that time period would struggle to understand and respond coherently to a question about a fictitious creature like the porridge bird laying an egg in the air. However, it’s worth noting that with advancements in natural language processing and machine learning techniques in recent years, modern language models are now able to understand and generate more complex language and even generate creative and imaginative responses.
…Did ChatGPT just jive me?
“…because if we’re lucky tomorrow, we won’t have to answer questions like yours anymore”